Module: Bolognese::Readers::SchemaOrgReader

Included in:
MetadataUtils
Defined in:
lib/bolognese/readers/schema_org_reader.rb

Constant Summary collapse

SO_TO_DC_RELATION_TYPES =
{
  "citation" => "References",
  "isBasedOn" => "IsSupplementedBy",
  "sameAs" => "IsIdenticalTo",
  "isPartOf" => "IsPartOf",
  "hasPart" => "HasPart",
  "isPredecessor" => "IsPreviousVersionOf",
  "isSuccessor" => "IsNewVersionOf"
}
SO_TO_DC_REVERSE_RELATION_TYPES =
{
  "citation" => "IsReferencedBy",
  "isBasedOn" => "IsSupplementTo",
  "sameAs" => "IsIdenticalTo",
  "isPartOf" => "HasPart",
  "hasPart" => "IsPartOf",
  "isPredecessor" => "IsNewVersionOf",
  "isSuccessor" => "IsPreviousVersionOf"
}

Instance Method Summary collapse

Instance Method Details

#get_schema_org(id: nil, **options) ⇒ Object



26
27
28
29
30
31
32
33
34
35
36
37
38
39
# File 'lib/bolognese/readers/schema_org_reader.rb', line 26

def get_schema_org(id: nil, **options)
  return { "string" => nil, "state" => "not_found" } unless id.present?

  url = normalize_id(id)
  response = Maremma.get(url)
  doc = Nokogiri::XML(response.body.fetch("data", nil), nil, 'UTF-8')

  # workaround for xhtml documents
  nodeset = doc.css("script")
  string = nodeset.find { |element| element["type"] == "application/ld+json" }
  string = string.text if string.present?

  { "string" => string }
end

#read_schema_org(string: nil, **options) ⇒ Object



41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
# File 'lib/bolognese/readers/schema_org_reader.rb', line 41

def read_schema_org(string: nil, **options)
  if string.present?
    errors = jsonlint(string)
    return { "errors" => errors } if errors.present?
  end

  read_options = ActiveSupport::HashWithIndifferentAccess.new(options.except(:doi, :id, :url, :sandbox, :validate, :ra))

  meta = string.present? ? Maremma.from_json(string) : {}

  identifiers = Array.wrap(meta.fetch("identifier", nil)).map do |r|
    r = normalize_id(r) if r.is_a?(String)
    if r.is_a?(String) && !r.start_with?("https://doi.org")
        { "identifierType" => "URL", "identifier" => r }
    elsif r.is_a?(Hash)
      { "identifierType" => get_identifier_type(r["propertyID"]), "identifier" => r["value"] }
    end
  end.compact.uniq

  id = normalize_id(options[:doi] || meta.fetch("@id", nil) || meta.fetch("identifier", nil))

  schema_org = meta.fetch("@type", nil) && meta.fetch("@type").camelcase
  resource_type_general = Bolognese::Utils::SO_TO_DC_TRANSLATIONS[schema_org]
  types = {
    "resourceTypeGeneral" => resource_type_general,
    "resourceType" => meta.fetch("additionalType", nil),
    "schemaOrg" => schema_org,
    "citeproc" => Bolognese::Utils::SO_TO_CP_TRANSLATIONS[schema_org] || "article-journal",
    "bibtex" => Bolognese::Utils::SO_TO_BIB_TRANSLATIONS[schema_org] || "misc",
    "ris" => Bolognese::Utils::SO_TO_RIS_TRANSLATIONS[resource_type_general.to_s.dasherize] || "GEN"
  }.compact
  authors = meta.fetch("author", nil) || meta.fetch("creator", nil)
  # Authors should be an object, if it's just a plain string don't try and parse it.
  if not authors.is_a?(String)
    creators = get_authors(from_schema_org_creators(Array.wrap(authors)))
  end
  contributors = get_authors(from_schema_org_contributors(Array.wrap(meta.fetch("editor", nil))))
  publisher = parse_attributes(meta.fetch("publisher", nil), content: "name", first: true)

  ct = (schema_org == "Dataset") ? "includedInDataCatalog" : "Periodical"
  container = if meta.fetch(ct, nil).present?
    url =  parse_attributes(from_schema_org(meta.fetch(ct, nil)), content: "url", first: true)

    {
      "type" => (schema_org == "Dataset") ? "DataRepository" : "Periodical",
      "title" => parse_attributes(from_schema_org(meta.fetch(ct, nil)), content: "name", first: true),
      "identifier" => url,
      "identifierType" => url.present? ? "URL" : nil,
      "volume" => meta.fetch("volumeNumber", nil),
      "issue" => meta.fetch("issueNumber", nil),
      "firstPage" => meta.fetch("pageStart", nil),
      "lastPage" => meta.fetch("pageEnd", nil)
    }.compact
  else
    {}
  end

  related_identifiers = Array.wrap(schema_org_is_identical_to(meta)) +
    Array.wrap(schema_org_is_part_of(meta)) +
    Array.wrap(schema_org_has_part(meta)) +
    Array.wrap(schema_org_is_previous_version_of(meta)) +
    Array.wrap(schema_org_is_new_version_of(meta)) +
    Array.wrap(schema_org_references(meta)) +
    Array.wrap(schema_org_is_referenced_by(meta)) +
    Array.wrap(schema_org_is_supplement_to(meta)) +
    Array.wrap(schema_org_is_supplemented_by(meta))

  rights_list = Array.wrap(meta.fetch("license", nil)).compact.map do |rl|
    hsh_to_spdx("__content__" => rl["name"], "rightsURI" => rl["id"])
  end

  funding_references = Array.wrap(meta.fetch("funder", nil)).compact.map do |fr|
    if fr["@id"].present?
      {
        "funderName" => fr["name"],
        "funderIdentifier" => fr["@id"],
        "funderIdentifierType" => fr["@id"].to_s.start_with?("https://doi.org/10.13039") ? "Crossref Funder ID" : "Other" }.compact
    else
      {
        "funderName" => fr["name"] }.compact
    end
  end
  dates = []
  dates << { "date" => meta.fetch("datePublished"), "dateType" => "Issued" } if Date.edtf(meta.fetch("datePublished", nil)).present?
  dates << { "date" => meta.fetch("dateCreated"), "dateType" => "Created" } if Date.edtf(meta.fetch("dateCreated", nil)).present?
  dates << { "date" => meta.fetch("dateModified"), "dateType" => "Updated" } if Date.edtf(meta.fetch("dateModified", nil)).present?
  publication_year = meta.fetch("datePublished")[0..3] if meta.fetch("datePublished", nil).present?

  if meta.fetch("inLanguage", nil).is_a?(String)
    language = meta.fetch("inLanguage")
  elsif meta.fetch("inLanguage", nil).is_a?(Object)
    language = meta.dig("inLanguage", 'alternateName') || meta.dig("inLanguage", 'name')
  else
    language = nil
  end

  state = meta.present? || read_options.present? ? "findable" : "not_found"
  geo_locations = Array.wrap(meta.fetch("spatialCoverage", nil)).map do |gl|
    if gl.dig("geo", "box")
      s, w, n, e = gl.dig("geo", "box").split(" ", 4)
      geo_location_box = {
        "westBoundLongitude" => w,
        "eastBoundLongitude" => e,
        "southBoundLatitude" => s,
        "northBoundLatitude" => n,
      }.compact.presence
    else
      geo_location_box = nil
    end
    geo_location_point = { "pointLongitude" => gl.dig("geo", "longitude"), "pointLatitude" => gl.dig("geo", "latitude") }.compact.presence

    {
      "geoLocationPlace" => gl.dig("geo", "address"),
      "geoLocationPoint" => geo_location_point,
      "geoLocationBox" => geo_location_box
    }.compact
  end

  # handle keywords as array and as comma-separated string
  subjects = meta.fetch("keywords", nil)
  subjects = subjects.to_s.split(", ") if subjects.is_a?(String)
  subjects = Array.wrap(subjects).reduce([]) do |sum, subject|
    sum += name_to_fos(subject)
    sum
  end

  { "id" => id,
    "types" => types,
    "doi" => validate_doi(id),
    "identifiers" => identifiers,
    "url" => normalize_id(meta.fetch("url", nil)),
    "content_url" => Array.wrap(meta.fetch("contentUrl", nil)),
    "sizes" => Array.wrap(meta.fetch("contenSize", nil)).presence,
    "formats" => Array.wrap(meta.fetch("encodingFormat", nil) || meta.fetch("fileFormat", nil)),
    "titles" => meta.fetch("name", nil).present? ? [{ "title" => meta.fetch("name", nil) }] : nil,
    "creators" => creators,
    "contributors" => contributors,
    "publisher" => publisher,
    "agency" => parse_attributes(meta.fetch("provider", nil), content: "name", first: true),
    "container" => container,
    "related_identifiers" => related_identifiers,
    "publication_year" => publication_year,
    "dates" => dates,
    "descriptions" => meta.fetch("description", nil).present? ? [{ "description" => sanitize(meta.fetch("description")), "descriptionType" => "Abstract" }] : nil,
    "rights_list" => rights_list,
    "version_info" => meta.fetch("version", nil).to_s.presence,
    "subjects" => subjects,
    "language" => language,
    "state" => state,
    "schema_version" => meta.fetch("schemaVersion", nil).to_s.presence,
    "funding_references" => funding_references,
    "geo_locations" => geo_locations
  }.merge(read_options)
end

#schema_org_has_part(meta) ⇒ Object



212
213
214
# File 'lib/bolognese/readers/schema_org_reader.rb', line 212

def schema_org_has_part(meta)
  schema_org_related_identifier(meta, relation_type: "hasPart")
end

#schema_org_is_identical_to(meta) ⇒ Object



204
205
206
# File 'lib/bolognese/readers/schema_org_reader.rb', line 204

def schema_org_is_identical_to(meta)
  schema_org_related_identifier(meta, relation_type: "sameAs")
end

#schema_org_is_new_version_of(meta) ⇒ Object



220
221
222
# File 'lib/bolognese/readers/schema_org_reader.rb', line 220

def schema_org_is_new_version_of(meta)
  schema_org_related_identifier(meta, relation_type: "SuccessorOf")
end

#schema_org_is_part_of(meta) ⇒ Object



208
209
210
# File 'lib/bolognese/readers/schema_org_reader.rb', line 208

def schema_org_is_part_of(meta)
  schema_org_related_identifier(meta, relation_type: "isPartOf")
end

#schema_org_is_previous_version_of(meta) ⇒ Object



216
217
218
# File 'lib/bolognese/readers/schema_org_reader.rb', line 216

def schema_org_is_previous_version_of(meta)
  schema_org_related_identifier(meta, relation_type: "PredecessorOf")
end

#schema_org_is_referenced_by(meta) ⇒ Object



228
229
230
# File 'lib/bolognese/readers/schema_org_reader.rb', line 228

def schema_org_is_referenced_by(meta)
  schema_org_reverse_related_identifier(meta, relation_type: "citation")
end

#schema_org_is_supplement_to(meta) ⇒ Object



232
233
234
# File 'lib/bolognese/readers/schema_org_reader.rb', line 232

def schema_org_is_supplement_to(meta)
  schema_org_reverse_related_identifier(meta, relation_type: "isBasedOn")
end

#schema_org_is_supplemented_by(meta) ⇒ Object



236
237
238
# File 'lib/bolognese/readers/schema_org_reader.rb', line 236

def schema_org_is_supplemented_by(meta)
  schema_org_related_identifier(meta, relation_type: "isBasedOn")
end

#schema_org_references(meta) ⇒ Object



224
225
226
# File 'lib/bolognese/readers/schema_org_reader.rb', line 224

def schema_org_references(meta)
  schema_org_related_identifier(meta, relation_type: "citation")
end


196
197
198
# File 'lib/bolognese/readers/schema_org_reader.rb', line 196

def schema_org_related_identifier(meta, relation_type: nil)
  normalize_ids(ids: meta.fetch(relation_type, nil), relation_type: SO_TO_DC_RELATION_TYPES[relation_type])
end


200
201
202
# File 'lib/bolognese/readers/schema_org_reader.rb', line 200

def schema_org_reverse_related_identifier(meta, relation_type: nil)
  normalize_ids(ids: meta.dig("@reverse", relation_type), relation_type: SO_TO_DC_REVERSE_RELATION_TYPES[relation_type])
end